Obstacles to linking emissions trading systems in the EU and China Zeng, Yingying
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Zeng, Y. (2018). Obstacles to linking emissions trading systems in the EU and China: A comparative law and economics perspective. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
6
LEAKAGE
AND
ETS
S
LINKING
3526.1 Introduction
The electricity sector353 has been the largest source of CO
2 emissions in both
the EU and China, and has been covered by both the EU ETS and China ETS.354
Therefore, reducing carbon emissions in the electricity sector is crucial to climate mitigation efforts, and electricity regulation (i.e. regulation of the electricity market and electricity emissions355) will be critical in determining whether or not the GHG
targets can be met. As mentioned in Chapter 4, this chapter will examine the electricity regulation in the China ETS and in particular the carbon leakage concerns it may generate.
352 This chapter builds upon a published article: Zeng et al., 2018. The authors would like to thank Joanna Depledge (Cambridge University, the UK) for her valuable comments and help with the publishing process. The authors are deeply grateful to Jingjing Jiang (Peking University, China), Suryapratim Roy (Trinity College Dublin, Ireland) and Fitsum Tiche (University of Groningen, the Netherlands) for their comments and suggestions. The authors would also like to thank the comments and advice from participants from 17th Global Conference on Environmental Taxation - Smart instrument mixes.
353 The electricity sector in this dissertation, as defined in Chapter 4.2.2, comprises sectors of generation, heat-power cogeneration and grid operators.
354 See State Grid Energy Research Institute and Yingda Media Investment Group, 2014; Eurostat, 2016; NDRC, 2016a; Olivier et al., 2016; NDRC, 2017.
355 As noted in Chapter 4, ‘electricity emissions’ generally refer to the GHG emissions released during the generation, transmission and distribution of electricity.
China has large spatial disparities between the primary energy resources that are used for electricity generation (e.g. coal) and the load centers.356 Some natural
resources like coal could be physically transported in large amounts to the load center but it may be rather costly, while some other resources (e.g. wind) cannot be easily transferred. Such a supply and demand mismatch in reality results in large parts of electricity to be transferred from the resource-rich areas to the load centers.357
In 2014, the overall amount of electricity transmitted inter-provincially/-regionally accounted for 20% of total electricity consumption and is expected to be growing.358
Admittedly, such inter-provincial/-regional electricity trade may lead to several environmental and economic benefits.359 But along with particular electricity flow,
a transfer of electricity generation from the ‘ETS regions’ to ‘less-regulated ETS regions’ (i.e. those with fewer carbon constraints) may give rise to a ‘leakage of electricity emissions’ within the electricity sector (hereafter ‘electricity carbon leakage’).360 The ‘less-regulated ETS region’ refers to the ETS region with less
stringent ETS rules (such as laxer carbon emission reduction goals and narrower coverage),361 potentially with lower carbon costs for similar covered entities than
356 See Kahrl et al., 2011.
357 See Lindner et al., 2013; China Electricity Council, 2015. 358 See China Electricity Council, 2015.
359 See Streets, 2003; Zhu et al., 2005; Wang, 2015.
360 Such a transfer of electricity generation and thus a certain electricity flow will give rise to carbon leakage, since ‘carbon leakage’ commonly refers to a shift of production and thus a transfer of carbon emissions from ‘carbon-constrained’ countries/regions/sectors to others without or with less carbon constraints (Directive 2003/87/EC, Art. 10a; Antimiani et al. 2013; Weishaar and Madani, 2014; Wang et al., 2017).
Specifically, carbon leakage can be driven via several channels. One that has been discussed in this chapter is through the ‘shift of production to other countries/regions with less carbon constraints (‘competitiveness discussion’). A second channel is through the ‘international markets’ of fossil fuel or other cleaner products (e.g. ethanol). For instance, a decreased demand for fossil fuel – as a result of carbon regulation – bring down the international prices of fossil fuels, which in turn induce an increased demand of fossil fuels in the less-constrained countries/regions. Additionally, carbon leakage may derive from policy instruments implemented at different levels (e.g. between the federal and state climate efforts). See, e.g., Goulder and Stavins, 2011; Zhang, 2012. 361 As analyzed below in Section 5.3.2, despite uniform standards to be applied (for the coverage and
allocation) in the national ETS, national carbon regulation also allows for more stringent rules to be implemented locally (provided being approved by the national government). See Arts.7 and 12 in NDRC, 2014a; Arts. 5 and 9 in NDRC, 2016b.
6
those in the ‘ETS region’. Accordingly, electricity carbon leakage, if any, is very likely to undermine the environmental effectiveness of the China ETS.
As analyzed above, in view of the potential gains from a future EU-China ETS linkage, political willingness and the ongoing bilateral cooperation on carbon markets, a linkage between the EU ETS and China ETS may materialize in the future. If such a linkage is to happen, environmental, efficiency and competitiveness implications of China’s electricity regulation and the potential electricity carbon leakage for its own system and further for its linking partner (the EU ETS) remain critical and thus merit further attention in the literature. For instance, in the eventuality of an EU-China linkage, ‘double counting of electricity emissions’362 (an
important regulatory feature in the China ETS) prima facie causes environmental and competitive concerns. Moreover, potential electricity carbon leakage in the China ETS appears to give competitive advantages to China’s large electricity users (mainly in the industrial sectors) and thus places different abatement cost burdens on equal entities in two systems (competitiveness concerns and efficiency losses).
The literature on the carbon leakage focuses largely on the ‘relocation of industrial production capacity abroad’ and corresponding compensation363 for the
industrial sectors in the EU ETS to, inter alia, sustain competitive advantages.364
While carbon leakage in the industrial areas is widely recognized, little attention in
362 ‘Double counting of electricity emissions’ refers to the fact that electricity emissions are double counted at both generation side and consumption side, which is to be analyzed in detail in sub-section 6.2.2.
363 For instance, the financial measures adopted in favour of sectors or subsectors determined to be exposed to a significant risk of carbon leakage due to costs relating to GHG emissions passed on in electricity prices; installations in sectors or subsectors which are exposed to a significant risk of carbon leakage shall be allocated allowances free of charge over phase III. See article 10a(6)(12) of Directive 2003/87/EC.
364 See Kuik and Hofkes, 2010; Monjon and Quirion, 2011; Branger et al., 2013; Juergens et al., 2013; Martin et al., 2013; Paroussos et al., 2015; Wang et al., 2017
practice or the literature has been given to electricity carbon leakage.365 Also,despite
the attention China’s electricity regulation and ETSs received thus far,366 very few
studies discuss in detail the electricity regulation (e.g. the inter-regional electricity trade) or double counting of electricity emissions in the context of the national ETS.367
Given the gap in the literature and potential EU-China ETS linkage, this chapter addresses the research question whether and how the inter- provincial/-regional electricity regulatory framework will give rise to electricity carbon leakage in the China ETS, and how it will impact the EU ETS once the world’s two largest systems are linked. To address the research question, this chapter employs a Law & Economics Approach to better understand ‘electricity carbon leakage’ that rests upon ‘economic incentive structures’ and equally complex ‘regulatory framework for electricity’. Specifically, under the current electricity regulation, electricity carbon
365 See Weishaar and Madani, 2014.
Marriot and Matthews (2005) developed a similar model for calculating the emissions embodied in electricity transfer between various states in US. Van den Bergh et al. (2014) formulated policy lessons in the EU ETS from the carbon leakage discussion in Regional Greenhouse Gas Initiative (RGGI) and the factors that give rise to the electricity leakage concerns, namely the ‘physical opportunity’ and ‘financial incentives’ for the power importation. Weishaar and Madani (2014) examined the evidence of the electricity leakage in the EU ETS based on an analysis of the inconsistencies between the EU ETS and energy market integration.
In the context of China, few papers that discussed the emissions spillover during the inter-regional electricity flow generally employed economic simulation. Meng et al. (2011) calculated the CO2
emissions for the electricity sector in each province, but from a consumption perspective by including the inter-regional electricity transfer between the six electricity grids, in which the grid average fuel mix was used to calculate the emissions of import/export between provinces. Meng et al. (2013) applied an inter-regional input-output model to explain the China’s inter-regional spillover of CO2 emissions, but from a perspective of domestic production network for 2002 and
2007. Lindner et al. (2013) developed a bottom-up model to calculate the direct CO2 emissions
embodied in electricity export and import between Chinese provinces. In addition, Zhu et al. (2005) discussed the environmental impact of the inter-grid connection in China, but primarily from the aspect of SO2 emissions.
366 See, e.g., Pang and Duan, 2016; Zeng et al., 2016b; Wang and Zhang, 2017. 367 See Li, 2012; Grubb et al., 2015; Wang et al., 2015; Zhang, 2015; Lin et al., 2016.
Particularly, regarding the electricity sector in the China ETS, Xun (2011) developed a model to identify the theoretical impacts of emissions trading on the electricity market and generators. Also, Li et al. (2012) applied a China State Information Center’s computable general-equilibrium model (SIC-GE) to analyze the short-term impact of carbon pricing, and found that the electricity sector is the most sensitive sector to the carbon price signal in China. Besides, Li (2012) pointed out several benefits of the inclusion of indirect emissions (associated with the purchased electricity) in the China ETS.
6
leakage arises when stakeholders (generators, consumers and grids) are incentivized into the most economically efficient decisions for participating in the inter-regional electricity trade and facilitating particular inter-regional electricity flow. Further, environmental effectiveness, efficiency and competitiveness in the combined systems will be examined to assess linking implications of electricity leakage.
The chapter is structured as follows. Section 6.2 introduces the highly regulated intra-and inter-regional electricity markets and ‘double counting’ of electricity emissions in the China ETS. Section 6.3 identifies the electricity leakage in the China ETS (i.e. two particular forms of inter-regional electricity flow) based on an analysis of the incentive structure of stakeholders. Linking implications of China ETS’s electricity carbon leakage are examined in Section 6.4. Section 6.5 summarizes the main conclusions.
6.2 Electricity regulation and double counting in
China
Two of the most striking features of electricity regulation in China are introduced in this section, namely the heavy government regulation of electricity in China’s electricity market (sub-section 6.2.1) and the ‘double counting of electricity emissions’ in the China ETS (sub-section 6.2.2).
6.2.1 Highly regulated electricity market in China
This sub-section analyzes how both intra-and inter-regional electricity markets are strongly regulated based on an introduction of the market structure and electricity pricing mechanisms.
There is no significant competition in the generation, transmission and the retail market. The 2003 power sector reform focused on the central-government-owned power plants. State Power Corporation (previously monopoly in the generation, transmission and distribution) was split into five generating companies, two grid companies, and several service companies.368 The five state-owned generators
account for most of overall power generation capacity,369 while the transmission and
368 See Wu, 2015.
369 See Polaris power grid, 2015a. The five state-owned generators accounts for a decreasing-but-dominating share of overall power generation capacity, respectively 47.8% in 2012, 46.5% in 2013 and 45.3% in 2014.
distribution remain under the ownership of the grid operators. Specifically, in China, there are seven inter-provincial regional networks (Northeast, Northwest, North, East, Central, South, and Guangdong) and five provincial networks (Shandong, Fujian, Xinjiang, Hainan, and Tibet). Those regional networks are run by two main state-owned grid companies including the State Grid Corporation of China (SGCC) and China Southern Power Grid (CSG).370 They each have a monopoly
in their respective regions: the CSG controls the South and Guangdong, and the much larger State Power Grid controls all the others. The State Grid is responsible for interregional dispatch, while each regional grid subsidiary has its own dispatch center.371
In March 2015, the State Council of the People’s Republic of China (hereafter ‘State Council’) released a new framework that proposed general reforms pushing forward on competition in the electricity generation and retail.372 Further, specifically
for the inter-provincial/-regional electricity transactions, the ‘Notice on improving the price formation mechanism for the inter-provincial/-regional electricity trade’ by NDRC, issued in May 2015, stipulates that both the sending and receiving parties may renegotiate the prices approved by government.373 While a comprehensive power
sector reform may be on its way, the reality in China suggests that this will not come any time soon.374 Currently, some pilots are slowly pushing forward reform in the
electricity market. For instance, Shenzhen and western Inner Mongolia have carried out a reform in the formulation of the transaction/distribution prices.375 It is also
expected that a nation-wide reform will build upon the experience gathered from pilots. Until this long-awaited reform is undertaken, this chapter focuses on the current intra-and inter-regional regulatory framework.
370 See Wu, 2015.
In 2011, the two grid companies own more than 95% of overall national installed/transmission capacity. In rural areas, some independent companies may have distribution networks where the national grid does not reach. See Pittman and Zhang, 2010.
371 See Zhou et al., 2010; Song et al., 2013; Wu, 2015.
372 See China Energy Storage Network, 2015; State Council of China, 2015. 373 See NDRC, 2015c.
374 See Zhang, 2015a.
6
1) The mandatory electricity prices in the intra-regional
electricity market
Generators face an on-grid tariff (price for electricity sold by power generators to the grid) set by government using benchmark pricing for both coal-fired electricity and renewable electricity. On the one hand, on-grid rates for coal-fired electricity have been set using benchmark pricing since 2004, in which generators in the same technology class are given the same tariff, based on an estimate of annual output and fixed and variable costs for that class.376 On the other hand, the on-grid tariff
for renewable electricity are set using regional benchmark prices (‘categorized on-grid pricing mechanism’), except that rates for hydropower are set mainly adopting ‘benchmark pricing’ along with other two approaches: 1) pricing on a facility-by-facility basis; 2) ‘receiving-end backward pricing mechanism’ (introduced in 2014) whereby the on-grid tariff is set in reference to prices in the electricity-receiving region.377
Consumption prices are also set by government and demonstrate regional differences.378 Local provinces determine the price ladder by taking into account
regional factors such as availability of natural resources and consumers’ affordability. Also, the consumption price in the same region differs among different categories of electricity end use (residential consumption, industrial & commercial consumption, and agricultural consumption), varied time periods of electricity use, various supply voltage and electricity capacity.379 Further, consumers – as long as they belong to
the same category of electricity end users within the same regional grid – pay the same consumption price to the local grid companies (or electricity-distributing companies) for electricity of the same voltage and capacity level,380 regardless of the
geographical generation locations of electricity. In other words, there is no price difference between the imported electricity and that locally generated for those consumers.
376 See Kahrl et al., 2011; Wu, 2015. 377 See Polaris power grid, 2015b.
378 See Xinhuanet, 2008. A striking characteristic of electricity prices in China is demonstrated that ‘prices in the south are higher than the north, prices in the east higher than the west’.
379 See Xinhuanet, 2012; NDRC, 2013. For instance, the mechanism of price ladder in residential electricity consumption has been implemented throughout the country since 2012, and residential consumption price generally increases when the amount of consumed electricity increases. 380 See National People’s Congress, 2015, Art. 41.
2) The strongly regulated inter-regional electricity trade
Most of the inter-regional electricity transactions are highly regulated to secure stable supply of electricity and also keep prices within acceptable bounds.381 In
particular, two state-owned grid companies (the State Grid and South Grid) remain in charge of the inter-regional electricity transactions and dispatch in their respective jurisdictions.382 To do so, the two grid companies develop the ‘Annual plan for the
inter-provincial/-regional electricity trade’ (hereafter ‘Annual plan’) in cooperation with the generating companies.383 For instance, at the beginning of each year, the
State Grid will issue the ‘Annual guidance plan for the inter-regional electricity trade’ (hereafter ‘Annual guidance plan’), requiring the provincial branches to incorporate the inter-regional electricity transactions (indicated in this guidance plan) into the ‘Provincial power balance arrangement’ and sign the legally binding contracts for purchases/sales. In this way, the Annual guidance plan may de facto turn into an inflexible and binding plan.
Further, electricity-pricing mechanisms differ to varied participants and different categories of electricity sales. On the one hand, the Inter-regional and Inter-provincial Electricity Trade Regulation stipulates that the electricity sellers include the licensed generating companies and the network entrusted by generators (e.g. State Grid or South Grid), while the electricity purchasers are the provincial grid companies, the qualified electricity-distributing companies and ‘independent consumers’ (i.e. those large users that are qualified to directly purchase electricity from the generators).384
On the other hand, the inter-regional electricity trade can be categorized into three main types, namely 1) the ‘point to network’ sales which refer to the direct sales of electricity from the qualified power plants (mainly in the power bases) to the regional/provincial grid companies; 2) the ‘network to network’ sales which refer to those transactions between provincial grids; 3) the ‘point to point’ sales, i.e. those transactions between the generators and independent consumers.385
381 See Tan, 2011; Zheng, 2011; State Electricity Regulatory Commission, 2012; Zhang, 2013; Wang, 2015.
382 See State Electricity Regulatory Commission, 2003, Art. 7. 383 See id., Art. 20.
384 See State Electricity Regulatory Commission, 2012, Art. 3; State Electricity Regulatory Commission, 2009.
6
The first two dominating categories of transactions, namely the ‘point to network’ and ‘network to network’ sales, implemented transaction prices that are mainly mandated by government. If prices have not yet been mandated, they can be set through negotiations between participants with reference to both the ‘average on-grid tariff’ at the sending end and the ‘average purchase price’ at the receiving end. Also, the transaction prices and volume in the ‘point to point’ transactions
shall be determined through negotiations between participants of electricity transactions, and grid companies shall charge ‘transmission costs’ that are mandated by government. Additionally, a portion of inter-regional electricity transactions
are conducted on the electricity-trading platform and thus implement transaction prices through market competition.
6.2.2 Double counting of electricity emissions in the China
ETS
In China, carbon costs and thus the carbon price signal cannot be passed from the side of electricity generation to the electricity users since the electricity consumption price is currently regulated.389 To incentivize consumers to reduce electricity use, the
regulators require that both the electricity generators and the electricity consumer must surrender emission allowances for the electricity that is being produced and consumed (double counting). On the one hand, emissions released from the
386 See NDRC, State Electricity Regulatory Commission and National Energy Board, 2009. 387 See ibid.
388 See Zhang, 2013; Li, 2015. 389 See Li, 2012; Zhang et al., 2014.
electricity generation and transmission are respectively counted as the emissions at the electricity generators and grids, i.e. ‘direct electricity emissions’.390
On the other hand, similar emissions are (double) counted as ‘emissions associated with the purchased electricity’ at the electricity consumers that are covered by the China ETS, i.e. ‘indirect electricity emissions’.391 Consequently, not only will
the generators and grids be incentivized to cut emissions (by adopting, e.g., more energy-efficient generation technology or clean generating fuels), but the electricity users (e.g. industrial users) are also incentivized to reduce electricity use.392
By contrast, in the EU ETS, only ‘direct electricity emissions’ are covered393
and allowances are auctioned to the installations of electricity generation.394 Since
390 See NDRC, 2013-2015.
‘Direct emissions’ are emissions from sources that are owned or controlled by the reporting entity, while ‘indirect emissions’ are emissions that are a consequence of the activities of the reporting entity, but occur at sources owned or controlled by another entity. See Greenhouse Gas Protocol, http://www.ghgprotocol.org/calculation-tools/faq (last accessed on Jan. 20, 2018).
Specifically, three broad scopes of power emissions can be categorized. Scope 1: all direct GHG emissions; scope 2: indirect GHG emissions from the consumption of purchased electricity, heat or steam; scope 3: other indirect emissions not covered in scope 2, such as the emissions associated with electricity-related activities (e.g. electricity losses during transmission & distribution, hereafter ‘T&D losses’) and the production of purchased materials and fuels. In this dissertation, for convenience purposes, emissions related to the T&D losses are considered as ‘direct emissions’ (at the electricity supply side) to contrast with ‘indirect emissions’ (at the electricity consumption side).
Altogether, ‘direct electricity emissions’ in this dissertation refer to emissions associated with certain amount of electricity from its generation, transmission and distribution, whereas ‘indirect electricity emissions’ are calculated when the same amount of electricity is consumed.
391 See NDRC, 2014a, Art. 47. 392 See Li, 2012; Zhang et al., 2014.
393 In the EU ETS, the operator shall assign all emissions from the combustion of fuels at the installation to the installation, regardless of exports of heat or electricity to other installations. The operator shall not assign emissions associated with the production of heat or electricity that is imported from other installations to the importing installation. See Annex IV in Commission Regulation (EU) No 601/2012 of 21 June 2012 on the monitoring and reporting of greenhouse gas emissions pursuant to Directive 2003/87/EC of the European Parliament and of the Council Text with EEA relevance.
394 According to Art. 10 of the revised ETS Directive 2003/87/EC (by Directive 2009/29/EC), from 2013 onwards, Member States shall auction allowances which are not allocated free of charge in accordance with Article 10a and 10c. This means in the third phase of the EU ETS (2013–2020), auctioning becomes the default method for allocating allowances and the share of allowances auctioned will rise progressively each year. See http://ec.europa.eu/clima/policies/ets/ index_en.htm, last accessed on 1st February 2015.
6
generators have to purchase allowances to cover their emissions, they increased the electricity prices to pass the carbon costs to electricity consumers.395 In this case, the
EU ETS inflates the costs of electricity consumption, because consumers have to pay higher electricity bills and thus ‘indirect carbon costs’.396 As a result, electricity
consumers are incentivized to reduce electricity use in the EU, even though they are not required to surrender allowances covering electricity emissions in the EU ETS.
6.3 Electricity leakage implications in the China ETS:
evidence from the regulatory framework of
inter-regional electricity trade
By examining economic incentive structure of stakeholders in the inter-regional electricity trade, this section first introduces the background for the inter-regional/-provincial electricity flow (sub-section 6.3.1), among which two particular forms of inter-regional/-provincial electricity flow are then identified as the electricity leakage in the China ETS (sub-section 6.3.2).
6.3.1 Incentive structures for the inter-provincial/-regional
electricity flow
This sub-section analyzes how the stakeholders (e.g. generators, electricity consumers and grids) are economically incentivized to participate in the
inter-395 See, e.g., Sijm et al., 2006; Carbon Point, 2008; Frondel et al., 2012; Jouvet and Solier, 2013; Schröder, 2013; Woerdman et al., 2013; Bönte, 2015.
Specifically, there has been a quite heated political and public debate, (especially) when allowances (to power sector) were largely freely allocated. Specifically, power generators have made additional profits – generally referred to as ‘windfall profits’ – by incorporating these emissions rights as ‘opportunity costs’ in the calculation of the cost price and passing on to consumers (Woerdman et al., 2013).
Sijm et al. (2006) also presents some empirically estimated rates of passing through CO2 opportunity costs (of EU ETS) to power prices in Germany and the Netherlands. Point Carbon (2008) calculated that, during the period 2005–2007, pass-through levels vary between 75 and 100% in both Germany, the United Kingdom and Spain, between 0 and 75% in Italy, and between 45 and 65% in Poland. Jouvet & Solier (2013) also calculated, compared and analyzed the extent of CO2 cost pass-through to electricity prices in the EU using an empirical model. 396 See, e.g., Woerdman et al., 2009; Frondel et al., 2012; Schröder, 2013; Bönte, 2015.
It bears mentioning that the most electro-intensive sectors in the EU ETS may be compensated for increases in electricity costs resulting from the ETS through national state aid schemes. See Art. 10a(6) of Directive 2003/87/EC.
regional electricity trade and thus facilitate the electricity flow based on an analysis of technical and regulatory opportunities.
First, ‘technical opportunities’ generally refer to those geographical disparities, resources characteristics and technical issues/limitations of generation/transmission that may generate the need for inter-regional electricity flow. For one thing, as mentioned above, China has large spatial disparities between the load centers and energy resources as well as between regions with resources.397 Some natural
resources like coal could be physically transported to the coal-fired power plants, but resources for other types of electricity (e.g. hydropower or wind power) cannot be easily transferred,398 generating the demand for the inter-regional electricity
trade. For another, the territory of China spans several tens of longitudinal and latitudinal degrees,399 and climatic differences across the country may create time
inconsistency in electricity demand (e.g. increasing demand for electricity in cold winter) and supply (e.g. fluctuating wind and sunlight period), generating further demand for the inter-regional electricity trade. Accordingly, three nationwide power transmission corridors (namely the north, the center and the south corridors) were built to promote the power networks’ interconnection and transmit power from the west to the east,400 facilitating the inter-regional electricity flow to a certain extent.401
Second, ‘regulatory opportunities’ refer mainly to the legal rules and regulations as well as the administrative/institutional setup and operation pertaining to the inter-regional electricity trade. Certain regulatory aspects may give rise to both
397 See Kahrl et al., 2011.
398 For instance, each hydro project has its very own hydrological and geological requirements. In this sense, technical limitations of generation may also generate demand for the inter-regional electricity transfer.
399 See Zhou, 2002.
400 See Pittman and Zhang, 2010; Zhou et al., 2010.
401 It bears mentioning that the current limited transmission capability and the stability problems of transmission systems have curbed the magnitude of inter-grid transmission. See Zhou, 2002. pp. 7-8; Zhou et al., 2010, pp. 4303-4309; Kahrl et al., 2011, pp. 4034-4035; Cableabc, 2015; National Energy Board, 2015; Wu, 2015.
Admittedly, some new technologies could be adopted to improve the transmission capacity and system stability to a certain extent, e.g. compact power transmission technology, series/parallel compensation technology and high voltage controlled shunt reactor (see Zhou, 2002, pp. 8-10; Zhou et al., 2010, pp. 4309-4312). Still, the infrastructure costs of network integration and technical costs to address those capacity and stability issues may discourage the grids to facilitate the inter-regional electricity flow.
6
environmentally beneficial and non-environmentally beneficial inter-regional electricity flow. On the one hand, for instance, the potential implementation of ‘renewable electricity consumption quota’ may politically incentivize the provincial grid companies to import the renewable electricity and thus facilitate the environmentally beneficial electricity flow mainly for two reasons. First, they are obliged to fulfill such quota that is associated with local consumption of renewable electricity (hydropower excluded).402 Second, they have the authority to import renewable electricity when
developing the ‘Annual plan’ (see above, section 6.2.1) and will be prone to do so especially when the locally generated renewable electricity is not yet sufficient to meet the quota.
On the other hand, the above-mentioned mandatory plan for the inter-regional electricity trade (e.g. Annual guidance plan) may stipulate unnecessary transactions and thus potential ‘non-environmentally beneficial electricity flow’. This can be exemplified by a case given in the ‘Regulatory report on the order of electricity trade in the Central Grid (2014)’. Even though the electricity supply in the Central Grid
has already exceeded its demand between January and April 2013, a large number of electricity was still transferred inter-regionally into the Central Grid from Shanxi province (in the North Grid), simply because such transaction had been previously indicated in the ‘Annual guidance plan’.
Allowing for both technical and regulatory opportunities, ‘economic drivers’ (e.g. market characteristics, electricity prices) will ultimately stimulate both the electricity purchasers and sellers into the most economically efficient investment decisions, which will then be reflected in the inter-regional electricity transactions. On the one hand, the grid companies and ‘independent consumers’ could be incentivized to actively participate in the inter-regional electricity trade as ‘electricity
402 According to the current draft (approved in principle by NDRC in August 2014), the ‘renewable electricity consumption quota’ refers to the renewable electricity that has been locally consumed (hydropower excluded) including the renewable electricity imported but excluding the exported. In this sense, both provincial governments and grid companies – especially in the provinces/ regions where renewable electricity is not affluent – may be politically incentivized to curtail the exportation of renewable electricity but encourage its importation. See Arts. 2,3,5,6 in National Energy Board, 2012; see also 21st Century Business Heral, 2014; Chinese Industry Research Network, 2014.
403 See National Energy Board, 2014. 404 See ibid.
purchasers’.405 Since the government-mandated on-grid tariffs and consumption
prices demonstrate regional differences, the locally generated may sometimes cost more than the imported, i.e. when the local on-grid tariff exceeds the sum of the selling price (charged by the sending end in the inter-regional sales) and the transmission price (mandated by government). Accordingly, the grid companies – that mainly benefit as ‘electricity sellers’ in the intra-regional electricity trade from the difference between the regulated selling price and the purchase price – and those independent electricity consumers will both be incentivized to import cheaper electricity.
On the other hand, certain generators may be incentivized to participate in the inter-regional electricity trade as ‘electricity sellers’. For instance, qualified hydropower plants could benefit from a recently introduced pricing mechanism for hydropower, ‘receiving-end backward pricing mechanism’.406 According to this
new mechanism, the selling price for the hydropower – generated and sent inter-regionally/-provincially by newly built hydro-power plants (built after February 1st, 2014) – is the difference between the purchase price (i.e. price paid by the receiving end) and the transmission price (T&D line loss included).407 Since the purchase
price can be determined through negotiations between the sending and receiving ends (in reference to the average purchase price in the power-receiving area), the qualified hydropower plants could sell the hydropower at a higher price than would otherwise be achieved in the intra-regional electricity trade (i.e. the local on-grid benchmark tariff).408
405 As mentioned above, the ‘independent consumers’ are qualified to directly purchase electricity from the generators in the inter-regional ‘point to point’ sales, while other dependent consumers can only purchase electricity from the local grids but not directly from generators.
406 See NDRC, 2014b; Polaris power grid, 2015b. 407 See NDRC, 2014b.
408 See Polaris power grid, 2015b.
Despite the on-grid tariff in the intra-regional trade is mostly regulated, hydropower plants could receive a higher selling price in the inter-regional trade (through negotiations). In those cases the regulated on-grid tariff in the power-receiving area is often higher than the regulated local on-grid tariff in the power-sending area.
6
6.3.2 Evidence for the electricity carbon leakage in the
China ETS
This sub-section identifies two particular forms of inter-regional electricity flow as the electricity leakage in the China ETS – the electricity leakage of ‘direct electricity emissions’ at generators’ side and ‘indirect electricity emissions’ at electricity consumers’ side (mainly in the industrial sectors).
1) Inter-provincial leakage of direct electricity emissions
‘Carbon leakage of direct electricity emissions’ refers to the electricity emissions spillover when generators shuffle resources (e.g. by transferring the generation location or resources) from the ETS regions to the less-regulated ETS regions due to potential regional differences in the ETS rules.
Generators covered by the ETS are legally required to surrender allowances for ‘direct electricity emissions’ and, potentially, ‘indirect emissions’ associated with the purchased electricity/heat. Specifically, the ‘direct emissions’ of generators include the combustion and desulphurization emissions that are largely contingent on the generation techniques and fuels adopted. Even though the uniform MRV guideline is adopted by the NDRC, the stringency level in the coverage and allocation for generators may still differ across different regions (see Table 6-1). This is because provincial DRCs are legally allowed to adopt ‘expanded coverage’ (e.g. lower thresholds) or ‘more stringent allocation’ (provided being approved by NDRC) and hence may deviate from national standards.409
Table 6-1 Potential regional differences of the ETS rules in the Chinese electricity sector (Phase I)
Coverage threshold for annual emissions Allocation MRV: REEFs adopted for
‘indirect electricity emissions’a
(KgCO2/kWh) National rules - 26,000 tonnes
(or 10,000 tonnes of standard coal equivalent). - Base year: 2013-2015.
- Lower threshold allowed provided being approved by NDRC.
Benchmarking allocation formula (Combined Heat-and-Power (CHP) generation excluded): A =
- Bi: uniform carbon emissions benchmark (for particular category of power generating unit considered) adopted by NDRC.
- Fp: ‘adjustment coefficient’ adopted by provincial DRCs;
potentially demonstrating ‘regional
difference’ due to different ‘regional economic development’ and ‘industrial planning’ (see Section 3.1).
A: the amount of allowances allocated; N: the amount of generating units. Fl: the cooling adjustment coefficient; Qi: the generation output of the generating unit considered.
Varied by 6 regional grids: East China: 0.7035 North China: 0.8843 Northeast China: 0.7769 Central China: 0.5257 Northwest China: 0.6671 South China: 0.5271 Pilots Beijing - 5,000 tonnes (base year: 2009-2012) - Most likely remaining
below national entry threshold
(see Section 3.1).
0.8843 Tianjin - 20,000 tonnes (2009-2012)
Shanghai - 20,000 tonnes (2010-2011) 0.7035
Fujian - 5, 000 tonnes of standard coal (2013-2015)
Guangdong - 20,000 tonnes (2011-2012) 0.5271
Shenzhen - 3,000 tonnes (2009-2012)
Chongqing - 20,000 tonnes (2008-2012) 0.5257
Hubei - 60,000 tonnes of standard coal
equivalent (2010-2011) - Requiring harmonization.
Source: Allocation Plans in pilots (2013-2016); NDRC, 2014c; Ideacarbon, 2016a; Ideacarbon, 2016b; NDRC, 2016a; NDRC, 2016b; NDRC, 2016c; Ideacarbon, 2017a; NDRC, 2017; Wang et al., 2017. a The latest ‘average regional electricity emissions factors (REEFs)’, issued by NDRC in 2012, are
6
Table 6-1 Potential regional differences of the ETS rules in the Chinese electricity sector (Phase I)
Coverage threshold for annual emissions Allocation MRV: REEFs adopted for
‘indirect electricity emissions’a
(KgCO2/kWh) National rules - 26,000 tonnes
(or 10,000 tonnes of standard coal equivalent). - Base year: 2013-2015.
- Lower threshold allowed provided being approved by NDRC.
Benchmarking allocation formula (Combined Heat-and-Power (CHP) generation excluded): A =
- Bi: uniform carbon emissions benchmark (for particular category of power generating unit considered) adopted by NDRC.
- Fp: ‘adjustment coefficient’ adopted by provincial DRCs;
potentially demonstrating ‘regional
difference’ due to different ‘regional economic development’ and ‘industrial planning’ (see Section 3.1).
A: the amount of allowances allocated; N: the amount of generating units. Fl: the cooling adjustment coefficient; Qi: the generation output of the generating unit considered.
Varied by 6 regional grids: East China: 0.7035 North China: 0.8843 Northeast China: 0.7769 Central China: 0.5257 Northwest China: 0.6671 South China: 0.5271 Pilots Beijing - 5,000 tonnes (base year: 2009-2012) - Most likely remaining
below national entry threshold
(see Section 3.1).
0.8843 Tianjin - 20,000 tonnes (2009-2012)
Shanghai - 20,000 tonnes (2010-2011) 0.7035
Fujian - 5, 000 tonnes of standard coal (2013-2015)
Guangdong - 20,000 tonnes (2011-2012) 0.5271
Shenzhen - 3,000 tonnes (2009-2012)
Chongqing - 20,000 tonnes (2008-2012) 0.5257
Hubei - 60,000 tonnes of standard coal
equivalent (2010-2011) - Requiring harmonization.
Source: Allocation Plans in pilots (2013-2016); NDRC, 2014c; Ideacarbon, 2016a; Ideacarbon, 2016b; NDRC, 2016a; NDRC, 2016b; NDRC, 2016c; Ideacarbon, 2017a; NDRC, 2017; Wang et al., 2017.
Further, in practice, different levels of stringency in the ETS rules (such as coverage and allocation rules) can be materialized as follows. On the one hand, the coverage threshold (for annual carbon emissions) determines whether a firm is covered by the ETS, since the China ETS regulates at the firm level not the installation level. With different coverage thresholds among different regions, similar entities may be covered in one region but not in another. Accordingly, generators can avoid abatement obligations by transferring the generation location to another region with a higher coverage threshold. For instance, evidence shows that most of the pilots (Hubei excluded) have already adopted lower coverage thresholds (Table 6-1). It is less likely that these pilots will simply abandon the previous coverage rules after they are included in the national ETS. This is mainly due to their climate ambitions and, more importantly, a lower national coverage-threshold later on that has been officially confirmed.410 In this case, the currently different coverage
thresholds among pilots – e.g. from 3, 000 tonnes of CO2 in Shenzhen to 20,000 tonnes of CO2 in Shanghai/Guangdong (see Table 6-1) – could shed a light upon the potential magnitude of future regional differences on this matter.
On the other hand, despite a uniform allocation method that is applied nationally for the power sector (e.g. emissions benchmark Bi, see Table 6-1), provincial DRCs may be incentivized to apply different ‘adjustment coefficients’ to the allocation (‘Fp’, see Table 6-1) for policy considerations such as ‘regional economic development’ and ‘industrial planning’.411 Such variations in ‘adjustment
coefficients’ may give rise to different stringency levels of allocation and thus lead to different amounts of allowances allocated for similar covered entities. Accordingly, by shifting the generation to the less-regulated ETS regions, the relocated generators could receive more allowances and have their abatement costs reduced.
Altogether, potential variations in the ETS rules could place different carbon cost burdens upon comparable covered generators in different regions. Given that the electricity sector is the most sensitive sector in China to the carbon cost signal,
generators in the ETS regions may be incentivized to reduce carbon costs simply by shuffling resources (without further abatement) to the less-regulated ETS regions. This can be realized in three main ways.
410 See Ideacarbon, 2016c; Ideacarbon, 2016d. 411 See NDRC, 2016c; Ideacarbon, 2017a. 412 See Li et al., 2012.
6
The first channel is to transfer the geographical location of generation. But in practice, this could be of limited influence since carbon costs may not be the primary concern when it comes to selecting a generation location. Also, relocating the generation cannot be easily done for certain types of power plants (e.g. hydropower, wind power), due to the geological requirements or technical difficulty of transferring energy resources. Moreover, regulatory obstacles such as the local protectionism of the government may discourage inter-regional transfer of large taxpayers (e.g. coal-fired plants).
A second way is to transfer the generation of resources (i.e. production input for generation) such as coal and gas. Still, this could be of limited relevance in practice since physically transporting resources for generation in large amounts and over long distance could be rather costly.
In light of the obstacles to a transfer of generation or resources, the third and most likely channel for the generators is through a simple shift in the generation output between different provinces/regions. For instance, for the trans-provincial/-regional generating companies, they can shuffle resources merely by increasing the electricity generation output in one area (the less-regulated ETS regions) while reducing the output in another without a physical transfer of generation location/ resources. Accordingly, those generators, ceteris paribus, could have their covered emissions reduced (higher coverage threshold) or receive more allowances (more lenient allocation) without substantial abatement (emissions leakage).
Consequently, with the ‘leakage of direct electricity emissions’ to the less-regulated ETS regions, more allowances will be rendered available in China’s system despite no further abatement (additionality/hot-air issues and environmental integrity concerns). Accordingly, those additional allowances will bring down the carbon price and thus discourage the abatement incentives in the system, undermining environmental effectiveness. In the long term, dynamic environmental effectiveness of the China ETS will also be jeopardized, with carbon leakage embodied in the above-mentioned generation-capacity transfer and a following decline of ‘actual abatement’ in the system.
2) Inter-regional leakage of indirect electricity emissions
As defined above, ‘direct electricity emissions’ in this dissertation refer to the emissions associated with a certain amount of electricity from its generation, transmission and distribution, whereas ‘indirect electricity emissions’ are calculated
when the same amount of electricity is consumed. For electricity consumers falling under the ETS, the ‘indirect electricity emissions’ are calculated by multiplying the amount of purchased electricity with a corresponding regional electricity emissions factor (REEF) that diff er across regions.413 Since indirect emissions are calculated in
one regional grid with one uniform REEF, ‘leakage of indirect emissions’ arises from the demand side on a regional level (not provincial level). Accordingly, diff erent indirect emissions calculated associated with similar amount of electricity may give rise to leakage, along with a fl ow of electricity from regions with high REEFs to regions with low REEFs. In 2011, approximately 35 billion kWh of electricity was transferred in this direction.414
Th ere is still some uncertainty about how the REEFs will be determined. Th e national MRV guideline only stipulates that REEFs adopted are those issued in the
413 See NDRC, 2013-2015.
414 See China Electricity Council, 2012.
Fig. 6-1 Electricity fl ow from regions with high REEFs to regions with lower REEFs in China (2011)
6
most recent year by the national authority (i.e. NDRC). But it is still unknown which REEF will be applied for the national ETS. Currently there are two REEFs that are being used in different pilots and one of them is therefore likely to be adopted in the China ETS.415
One REEF is the ‘benchmark emissions factors for the regional grids’ (hereafter the ‘benchmark REEF’) issued by the NDRC.416 It was originally designed for CDM/
CCER projects development417 but has been used in practice to calculate the indirect
emissions associated with the electricity consumption (e.g. in the Shenzhen Pilot).418
The benchmark REEF does not fully consider the environmental effect of regional electricity importation.419 Moreover, it does not consider the renewable electricity.
The other one is the ‘average CO2 emissions factors for the regional grids’ (hereafter the ‘average REEF’) issued by the National Center for Climate Change Strategy and International Cooperation (NCSC). This center is affiliated to the NDRC. According to NDRC, the average REEF can be used to calculate the indirect electricity emissions.420 Its calculation considers the net importation of electricity
from other regional grids or countries and their electricity emissions factors (see
415 For instance, the ‘benchmark REEF’ is adopted in the Shenzhen pilot as the default electricity emission factors (specifically, the EFgrid,OMsimple in the South Grid in the ‘2009 - 2011 China
benchmark emissions factors in the regional grids’). See p. 29 in ‘Specification with guidance for quantification and reporting of the organization’s greenhouse gas emissions (SZDB/Z 69 - 2012)’, issued on Nov. 6, 2012 by Market Supervision Administration of Shenzhen Municipality. The 2010 ‘average REEF’ is adopted in the Tianjin pilot (specifically, the average REEF on the
Tianjin municipality level but not on the regional level). See Table B-3 ‘Default emission factors for the purchased electricity/heat’ in ‘Tianjin MRV guideline’.
Additionally, the Shanghai pilot applies neither of them. Instead, the default factors are calculated according to ‘Shanghai 2010 energy balance table’ and ‘Greenhouse gas inventory’. See p 12 in Shanghai GHG emissions MRV Guidelines (SH/MRV-001-2012) issued on Dec. 11, 2012 by Shanghai Municipal DRC.
The REEFs issued by the regional grids – despite being widely used – are not discussed since they are not calculated by the authority and thus less likely to be employed in the national ETS. 416 For the detailed calculation of the benchmark REEF, see http://cdm.ccchina.gov.cn/zyDetail.
aspx?newsId=52507&TId=160. 417 See NDRC, 2015e, para.1. 418 See Song et al., 2013; Lv; 2014.
419 In case of electricity exports to the regions with a higher benchmark REEF, there would of course be opposite effects on the environment.
formula 1).421 The ‘average REEF’, however, only takes into account the combustion
emissions from fossil fuels (non-fossil fuels excluded) and thus far is mostly issued with considerable delay (the average REEFs for 2011/2012 were issued only in September 2014).422
Average REEF
EFgrid, i = (1)
Emgrid, i = ∑m (FCm NCVm (2)
Note: EFgrid,i, EFgrid,j and EFk are the average electricity emissions factors in regional grid i, another regional grid j and another country k. Egrid,i is the annual aggregate generation volume within the geographical scope of regional grid i, while Eimp,j,i and Eimp,k,I represent the net imported electricity volume from regional grid j and the country k. Emgrid,i refers to the combustion emissions from all the fossil fuels (during the generation) within the grid i.
Whichever REEF will be adopted, either of them can give rise to electricity carbon leakage.
The benchmark REEF does not fully consider the carbon content of (regional) electricity imports as it applies the local regional REEF423 irrespective of the actual
carbon content of electricity imports. Consequently, the factual indirect electricity emissions will be under-calculated (‘importation effects’) in case of electricity imports from regions that have a higher REEF benchmark and hence give rise to electricity leakage and environmental integrity concerns.424
421 In the formula 1, when the local regional grid i imports electricity from another grid j (i.e. Eimp,j,i),
since the emissions factor for the imported electricity (EFgrid,j) has already been considered into
the calculation of the ‘average REEF’ (EFgrid,i), according GHG emissions (associated with the
imported electricity) produced during the generation (‘importation effects’) will be accounted as well, see ∑j(EFgrid,j×Eimp,j,i).
422 See NDRC, 2014c.
423 See NDRC, 2013-2015; Li, 2012; Song et al., 2013; Zhang et al., 2014.
424 REEF merely represents the average green level of generation in one region. Hence, it is likely that the imported electricity is de facto cleaner than the locally generated. Still, indirect electricity emissions are under-calculated on the whole in those cases, since the clean imported electricity has already been reckoned into the calculation of REEF in the power-exporting region.
6
Also, in the case of ‘average REEF’ indirect electricity emissions will be under-calculated, although the average REEF does take into account the carbon content of imported electricity (see formula 1 above). Indirect emissions will be under-calculated in the case of locally generated (Egrid,i) and the imported (∑jEimp,j,i) electricity. In principle, the direct electricity emissions calculated for the generators are equal to the indirect electricity emissions counted for the electricity consumers (line loss during transmission & distribution excluded).425 But the measurement of
electricity emissions in practice is not the case. In other words, different amount of emissions are counted between direct and indirect emissions associated with the same amounts of electricity. For one thing, the direct electricity emissions currently calculated the combustion emissions (only from the fossil fuels, e.g. biofuel excluded) during the generation and the CO2 emissions produced during the desulfurization process for the coal-fired power plants.426 For another, the ‘average REEF’ – if
applied to calculate indirect electricity emissions – considers only the combustion emissions from the fossil fuels (see Emgrid,i).427 Altogether, both direct and indirect
electricity emissions fail to include the combustion emissions from non-fossil fuels, and indirect emissions further leave out the CO2 emissions produced during the desulfurization process for the coal-fired power plants (‘under-calculation effects’).428
Still, the adoption of the ‘average REEF’ could give rise to electricity leakage concerns, for instance, when region i with a lower REEF (EFgrid,i) imports cheaper-but-dirtier electricity from region j (with a higher REEF).429 In other words, the dirtier
electricity is further generated in grid j and exported to replace or complement the electricity demand in region i (with cleaner local generation). Altogether, with such
425 As indicated before, emissions during the transmission and distribution (e.g. the line losses) are not discussed in this chapter for convenience purposes, since they accounted for a rather small percentage of electricity emissions.
426 See NDRC, 2013-2015.
427 The indirect emissions calculated for clean and dirty electricity do not differ (to electricity consumers). In this sense, there will be no substitution or guidance effects to incentivize those consumers to purchase clean electricity. This is not discussed since it is beyond the scope of the chapter (i.e. electricity leakage).
428 It bears mentioning that the electricity emissions leakage has yet to happen since such ‘under-calculated effects’ occur without the inter-regional electricity flow.
429 Generators in region j – if with lower carbon costs – would have carbon competitive advantage over region i and thus be able to offer cheaper electricity during the inter-regional trade.
a electricity flow, a further ‘under-calculation effects’430 (for the indirect emissions) –
associated with the imported electricity (i.e. Eimp,j,i) – will cause electricity leakage.431
To sum up, with both REEFs, this particular inter-regional electricity flow leads to the indirect electricity emissions under-calculated (electricity leakage). Accordingly, in the long run, the ratio of direct emissions calculated (to the indirect emissions) in the ETS will increase in both cases but for different reasons. On the one hand, in the case of the benchmark REEF, this ratio arises because the carbon content of the imported electricity is not fully counted and the indirect emissions for the imported are generally under-calculated (‘importation effects’). On the other, when the average REEF is adopted, indirect emissions for the imported electricity are under-counted (i.e. the ‘under-calculation effects’). When dirtier electricity is further generated to complement an increased demand in the region with cleaner generation, the ‘further under-calculation effects’ (identified above) with regard to the particular inter-regional flow will cause electricity leakage. Meanwhile, the proportion of direct emissions arises mainly because CO2 emissions from the desulfurization process are not considered.
As a result of such electricity flow, different effects may arise in the short and long run. In the short term, actual electricity emissions may increase compared to the scenarios without such flow (environmental effectiveness jeopardized), since the dirtier imported-electricity is used to replace or complement the cleaner locally generated electricity. Moreover, there will not be any change to the allowances surplus nor the carbon price. This is because generators in the China ETS are granted allowances using output-based allocation methods: benchmarking (see Table 6-1).432
When the local generators reduce the power output due to the upcoming electricity
430 The ‘under-calculation effects’ refers to the under-calculation of CO2 emissions released from
the non-fossil fuels combustion and desulfurization from the coal-fired power plants, when the average REEF is adopted to count the indirect emissions for the imported electricity. Hence, it bears mentioning such under-calculation effects do not occur when other fossil fuels (e.g. natural gas) are used for generation.
431 It bears mentioning the adoption of the ‘benchmark REEF’ would give rise to similar under-calculation effects (electricity leakage) since it does not consider the renewable electricity either. 432 An exception is that the ‘historical carbon intensity reduction method’ is adopted for CHP
6
importation and the electricity emissions decrease,433 accordingly they will receive
fewer allowances. The same goes for the generators in the electricity-exporting regions (i.e. increasing electricity emissions and a pro rata increase in the allowances). As a result, the surplus and carbon price remain unchanged since the supply and demand of allowances in the whole system have changed to the same extent within a certain range.434
In the long term, as analyzed above, this particular electricity flow increases the proportion of direct allowances (to the indirect allowances). Accordingly, the China ETS – originally covering both upstream and downstream systems – may gradually move towards an ‘upstream ETS’.435 First, more abatement burdens may be shifted
to generators (water-bed effects), possibly decreasing the political acceptability of the ETS (distributional effects) since generators in China’s system are under increasing abatement pressure and cannot pass on the carbon costs to the downstream grid or electricity-consumers. Second, an upstream ETS may bring more direct control and thus meet the environmental objective with a lower level of uncertainty.436 Such
uncertainty may partially arise from the fact that the current electricity prices are highly regulated and the carbon cost signal cannot be passed from the generators to consumers, resulting in generators’ failure to comply. Third, it may enhance the efficiency in the sense that the potential scale effects of abatement at the generators may bring down the aggregate abatement costs. In addition, transaction costs may
433 Since the Annual plan is oftentimes developed the year ahead, local generators will be notified of the upcoming electricity importation and adjust their generation output accordingly. Further, this chapter assumes that regulators will have enough time and capacity to incorporate the above-mentioned changes into the process of allocation.
434 It bears mentioning the supply of allowances in the China ETS can only increase to a certain extent due to the upper-limit of aggregate allowances (despite the intensity-based cap adopted in China). See Zeng et al. (2016).
435 In any industry, there is a vertical chain of production and consumption, with several ‘layers’ from initial production to final consumption (Kerr & Duscha, 2014, p 6). In terms of the electricity sector, an upstream ETS regulates only the emission sources (generators, e.g. in the EU ETS), while a downstream ETS regulates emissions relating to the consumption of electricity at the electricity consumers’ side. Since the Chinese regulators require that both the electricity generators and electricity consumers must surrender emission allowances for the electricity that is being produced and consumed (double counting), the China ETS covers both upstream and downstream system.
also be reduced because of the higher emissions per regulated entity and fewer regulated entities involved.437
6.4 Linking the China ETS to the EU ETS: implications
of China ETS’ electricity leakage for the EU
As analyzed above, an EU-China ETS, although only likely to materialize in the longer term, remains a crucial issue towards global climate change mitigation efforts that certainly merits further attention in the literature. Specifically, with a hypothetic linkage between both ETSs, this section analyzes the implications of China ETS’ electricity leakage for the EU ETS based on the analysis of two particular categories of electricity carbon leakage in section 6.3. Particularly, environmental effectiveness/ integrity, competitiveness and efficiency concerns are addressed.
To begin with, both types of leakage (of direct and indirect electricity emissions) in the China ETS will certainly undermine the environmental integrity of the linked ETSs but may jeopardize the environmental effectiveness of the EU ETS in different ways. As analyzed above, the ‘electricity leakage of direct electricity emissions’ on the supply side in the China ETS renders more allowances available in the China ETS and brings down the carbon price. In this case, more cheaper allowances – ceteris paribus – will flow from China’s system into its linking partner. Meanwhile, the abatement incentives in the EU ETS will most likely be discouraged due to the carbon price decrease, thus jeopardizing environmental effectiveness in the EU ETS.
By contrast, the ‘electricity leakage of indirect electricity emissions’ on the demand side in the China ETS will not change the allowances surplus or carbon price in China’s system, since in this case both granted allowances (supply) and covered emissions (demand) inflate to the same extent within a certain range. Accordingly, the environmental effectiveness of the EU ETS will not be immediately or directly undermined but may still be compromised especially in the long term – with the flow of dirtier electricity from the regions with higher REEF to other regions with lower REEF. This is because – with the current carbon regulatory framework – the China’s system and thus the linked systems will be granting increasing allowances to dirtier generators and encouraging cheap-but-dirty electricity to replace or complement the
6
cleaner electricity in those regions (with lower REEF), hence discouraging abatement incentives in the linked systems (dynamic environmental effectiveness jeopardized).
Second, electricity leakage in the China ETS may give rise to competitiveness concerns to the covered entities (industrial sectors) in the linked systems. Specifically, it may affect the competitive position of ‘independent consumers’ (normally large users in the industrial sectors) in China’s system but not of the ‘dependent consumers’. On the one hand, compared to scenarios without such flow, dependent consumers pay to the grid companies (or electricity-distributing companies) the same electricity consumption costs (that are mandated by government) and the same carbon costs associated with indirect electricity emissions (that are calculated with the same local REEF). On the other, since independent consumers are legally allowed to directly purchase electricity from generators in other provinces/regions, they are able to import dirtier electricity at a cheaper electricity price without shouldering higher carbon costs. As a result, competitive distortions may result in the sense that large electricity users in the industrial sectors (i.e. independent electricity consumers) gain a competitive advantage over their competitors in the EU ETS.
Furthermore, one prominent issue that certainly causes competitive and efficiency concerns is ‘double counting’ (the fact that electricity emissions are covered and measured at both generation and consumption side) in the China ETS. The very few papers available that discussed double counting have generally expressed concerns towards it.438 In the context of linking, the indirect emissions of
the purchased electricity are covered and measured in the China ETS but not in the EU ETS. It seems at first glance to be different coverage & MRV rules between the two systems. Since different coverage and MRV rules between the linked systems
438 Sorrell (2003a) pp. 692-696 identified ‘double counting’ of emission reductions as one problem raised by the coexistence of the EU ETS with the UK Renewables Obligation (RO) and the UK Energy Efficiency Commitment (EEC). Ellis & Tirpak (2006) p. 22 and Jakob-Gallmann (2011) p. 286 generally mentioned that ‘care and appropriate accounting procedures would be needed’ to avoid double counting and ensure environmental integrity of the system. Schneider et al. (2015) assesses how double counting can occur and how it could be addressed at the international level.
